[PATCH v2 2/8] lib: vdso: Build 32 bit specific functions in the right context

Vincenzo Frascino vincenzo.frascino at arm.com
Tue Jan 14 03:15:23 PST 2020


Hi Thomas,

On 14/01/2020 10:37, Thomas Gleixner wrote:
> Vincenzo Frascino <vincenzo.frascino at arm.com> writes:
>>
>> On 14/01/2020 09:33, Thomas Gleixner wrote:
>>> Thomas Gleixner <tglx at linutronix.de> writes:
>>>
>> [...]
>>
>>>
>>> Bah, it's not fixing it. That's what you get when you compile the wrong
>>> tree...
> 
> This part is required to cover the BUILD_VDSO32 guard, but then when the
> fallback thing is removed it fails again because the 32bit fallbacks are
> missing.
> 
> The patch below makes it build again.
I completed the testing and everything seems fine. For completeness I am
reporting the test results below:

clock-gettime-monotonic: syscall: 938 nsec/call
clock-gettime-monotonic:    libc: 278 nsec/call
clock-gettime-monotonic:    vdso: 270 nsec/call
clock-getres-monotonic: syscall: 678 nsec/call
clock-getres-monotonic:    libc: 692 nsec/call
clock-getres-monotonic:    vdso: 33 nsec/call
clock-gettime-monotonic-coarse: syscall: 840 nsec/call
clock-gettime-monotonic-coarse:    libc: 184 nsec/call
clock-gettime-monotonic-coarse:    vdso: 172 nsec/call
clock-getres-monotonic-coarse: syscall: 710 nsec/call
clock-getres-monotonic-coarse:    libc: 733 nsec/call
clock-getres-monotonic-coarse:    vdso: 35 nsec/call
clock-gettime-monotonic-raw: syscall: 894 nsec/call
clock-gettime-monotonic-raw:    libc: 278 nsec/call
clock-gettime-monotonic-raw:    vdso: 270 nsec/call
clock-getres-monotonic-raw: syscall: 669 nsec/call
clock-getres-monotonic-raw:    libc: 696 nsec/call
clock-getres-monotonic-raw:    vdso: 35 nsec/call
clock-gettime-tai: syscall: 933 nsec/call
clock-gettime-tai:    libc: 277 nsec/call
clock-gettime-tai:    vdso: 264 nsec/call
clock-getres-tai: syscall: 674 nsec/call
clock-getres-tai:    libc: 696 nsec/call
clock-getres-tai:    vdso: 33 nsec/call
clock-gettime-boottime: syscall: 934 nsec/call
clock-gettime-boottime:    libc: 278 nsec/call
clock-gettime-boottime:    vdso: 270 nsec/call
clock-getres-boottime: syscall: 677 nsec/call
clock-getres-boottime:    libc: 690 nsec/call
clock-getres-boottime:    vdso: 33 nsec/call
clock-gettime-realtime: syscall: 901 nsec/call
clock-gettime-realtime:    libc: 278 nsec/call
clock-gettime-realtime:    vdso: 272 nsec/call
clock-getres-realtime: syscall: 677 nsec/call
clock-getres-realtime:    libc: 701 nsec/call
clock-getres-realtime:    vdso: 33 nsec/call
clock-gettime-realtime-coarse: syscall: 838 nsec/call
clock-gettime-realtime-coarse:    libc: 184 nsec/call
clock-gettime-realtime-coarse:    vdso: 172 nsec/call
clock-getres-realtime-coarse: syscall: 713 nsec/call
clock-getres-realtime-coarse:    libc: 736 nsec/call
clock-getres-realtime-coarse:    vdso: 35 nsec/call
getcpu: syscall: 620 nsec/call
getcpu:    libc: 648 nsec/call
gettimeofday: syscall: 1022 nsec/call
gettimeofday:    libc: 280 nsec/call
gettimeofday:    vdso: 272 nsec/call

-- 
Regards,
Vincenzo
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pEpkey.asc
Type: application/pgp-keys
Size: 14071 bytes
Desc: not available
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20200114/ecb1112b/attachment.bin>


More information about the linux-arm-kernel mailing list